Readme file for Huebert Group cascade impactor mass data:

We operated 8-stage Micro-Orifice Uniform Deposit Impactors (MOUDIs) at Cape Grim during the ACE-1 project.  They were connected to the 10m inlet used by several other investigators.  It heated entering air to reduce the relative humidity to approximately 50%.

Samples were generally changed daily, though were occasionally extended to collect larger samples.  Some effort was taken to coordinate sample times with meteorological changes.  Sampling was continuous, whether or not baseline conditions existed.

The impactor samples that were weighed were all dedicated to organic analysis, rather than the inorganic analysis used in the other cascade impactor data set we submitted to CODIAC.  Therefore, while these samples were typically coordinated with the inorganic samples, they were _not_ the same samples.  To compare

In addition to the 8 impaction stages, we also analyzed the inlet stage and a backup filter.  Impaction substrates were aluminum foil that had been baked in an oven to eliminate organic material.  Baked glass fiber after-filters were used but shed fibers during handling and sampling, so the mass results were meaningless and are not included.

Exposed samples were dried in a dessicator for at leat 12 hours, then weighed in a microbalance housed in a chamber containing silica gel dessicant.  Relative humidity was estimated to be less than 15%.

The design flow rate for the MOUDI is 30 liters per minute.  Nominal cut diameters of the MOUDI stages are:

Stage    50% cut (aerodynamic diameter, microns)
----------------
inlet      15     (though the 10 m inlet efficiency is not known)
  1        10.
  2         5.6
  3         3.2
  4         1.8
  5         1.0
  6         0.56
  7         0.32
  8         0.18
filter      not well defined, though very small.  We assume 0.10 microns
            for plotting purposes only, as little mass is present below
            that size.

Aerodynamic diameter is defined as the size of a particle with unit density (1g/ml) with the same aerodynamic behavior (Stokes number) as a sampled particle.  For particles > 0.1 micron or so, this works out to approximately

diameter = aerodynamic diameter / sqrt(particle density).

Flow rates were not always 30 lpm, so the data file indicates corrected size cuts, determined by using

cut = nominal cut * sqrt(nominal flow/actual flow).

File format:

The data file is ASCII text.  Fields are separated with tab characters.

There are two header lines.  The first names each column, while the second gives units.

The columns are:
1)  Sample name.  This is included primarily to make it clear which substrates
    were exposed together in the same sample.
2)  Beginning day for the sample (day of year)
3)  Beginning time, HHMM  (local time, not UTC or station time)
4)  Ending day for the sample (day of year)
5)  Ending time, HHMM  (local time)
6)  Impactor stage (0 for inlet, 1-8, F for after-filter)
7)  50% cut size for the stage, corrected for flow rate (aerodynamic 
    diameter, microns).  As a very rough first approximation, one can assume 
    that each stage captures all particles larger than its size cut and smaller 
    than that of the previous stage.
8)  Mass concentration of particles captured on the stage, reported as
    nanograms per standard cubic meter of air.  Standard conditions were
    defined as 293 K and 101.325 kPa.
9)  Estimated error (95% confidence limits) based on blank variability.  Errors 
    in the flow rate, which affect all stages equally, are estimated at 10%, 
    and are not included.